한국ITS학회 논문지 (The Journal of The Korea Institute of Intelligent Transport Systems)

  • 제20권4호
  • /
  • Pages.46-56
  • /
  • 2021
  • /
  • 1738-0774(pISSN)
  • /
  • 2384-1729(eISSN)
한국ITS학회 (The Korea Institute of Intelligent Transport Systems)
권호 표지
DOI QR코드

DOI QR Code

돌발상황 처리시간 예측을 위한 영향요인 분석 및 SMOGN-DNN 모델 개발

Analysis of Incident Impact Factors and Development of SMOGN-DNN Model for Prediction of Incident Clearance Time

  • 윤규리 (부경대학교 공간정보시스템공학과) ;
  • 배상훈 (부경대학교 공간정보시스템공학과)
  • Yun, Gyu Ri (Dept. of Spatial Information Eng., Pukyong National University) ;
  • Bae, Sang Hoon (Dept. of Spatial Information Eng., Pukyong National University)
  • 투고 : 2021.06.22
  • 심사 : 2021.08.06
  • 발행 : 2021.08.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

돌발상황으로 인한 비반복정체로 발생하는 높은 교통비용과 혼잡을 효과적으로 해소하기 위해서 돌발상황 처리시간을 예측하는 것은 중요하다. 본 연구에서는 인공신경망을 활용한 예측모델 개발을 위해 국내 도로상황에 적합한 돌발상황 처리시간 영향요인을 분석하고, 이를 학습데이터로 생성하였다. 기존 연구에서 장시간 소요되는 돌발상황 처리시간에 대한 과소 예측 문제가 발생하여 이에 대한 해결방안으로 본 연구에서는 SMOGN기법을 적용한 오버샘플링 학습데이터를 생성하여 이를 모델에 적용하였다. 그 결과 SMOGN기법을 적용한 DNN모델이 MAE 18.3분으로 연구 과정에서 구축된 모델 중 가장 높은 정확도로 돌발상황 처리시간을 예측하여, 기존에 개발된 예측모델의 한계점을 보완할 수 있을 것으로 기대한다.

Predicting the incident clearance time is important for eliminating the high transportation costs and congestion from non-repetitive congestion caused by incidents. In this study, the factors influencing the clearance time suitable for domestic road conditions were analyzed, using a training dataset for predicting the incident clearance time using artificial neural networks. In a previous study, the under-prediction problem for high incident clearance time was used. In the present study, over-sampling training data applied using the SMOGN technique was obtained and applied to the model as a solution. As a result, the DNN model applying the SMOGN technique could compensate for the limitations of the previously developed prediction model by predicting the clearance time with the highest accuracy among the models developed in the research process with MAE = 18.3 minutes.

키워드

참고문헌

  1. Ha O. K.(2010), "The Prediction Models for Clearance Times for the unexpected Incidences According to Traffic Accident Classifications in Highway," The Journal of the Korea Institute of Intelligent Transportation Systems, vol. 9, no. 1, pp.101-110.
  2. Han W. G.(2001), "A Model For Estimating Incident Duration," Korean Society of Civil Engineers, vol. 21, no. 3-D, May.
  3. Kim J. W.(2005), Development of a model for estimating incident duration: Focusing on seoul urban expressway traffic management systems, University of Seoul.
  4. KoROAD(2021), Traffic Accident Analysis System, http://taas.koroad.or.kr
  5. Lee K. Y. et al.(2012), "A Study on the Influencing Factors for Incident Duration Time by Expressway Accident," Korean Society of Road Engineers, vol. 14, no. 1, pp.85-94.
  6. Lee S. B., Han D. H. and Lee Y. I.(2015), "Development of Freeway Traffic Incident Clearance Time Prediction Model by Accident Level," J. Korean Soc. Transp., vol. 33, no. 5, pp.497-507. https://doi.org/10.7470/jkst.2015.33.5.497
  7. Paula B., Luis T. and Rita P. R.(2017), "SMOGN: A Pre-processing Approach for Imbalanced Regression," Proceedings of the First International Workshop on Learning with Imbalanced Domains: Theory and Applications, PMLR, vol. 74, pp.36-50.
  8. Ruimin L., Francisco C. P. and Moshe E. B. A.(2018), "Overview of traffic incident duration analysis and prediction," European Transport Research Review, vol. 10, p.22. https://doi.org/10.1186/s12544-018-0300-1
  9. Shin C. H. and Kim J. H.(2002), "Development of Freeway Incident Duration Prediction Models," Journal of Korean Society of Transportation, vol. 20, no. 3, pp.17-30.
  10. The Korea Transport Institute(2019), Transportation cost calculation status and improvement plan.
  11. Valenti G., Lelli M. and Cucina D.(2010), "A Comparative study of models for the incident duration prediction," Eur Transp Res Rev., vol. 2, no. 2, pp.103-111. https://doi.org/10.1007/s12544-010-0031-4